Inexpensive Ballscrews and Ballnuts?

[firestorm1284]

Hey everyone, still looking for more ballscrew info. I'm doing an X3 conversion and am definitely going to be using ballscrews. I want to keep the accuracy as high as reasonably possible. So far it seems that will involve a rolled ballscrew instead of a ground one for cost reasons, and also fabricating preloaded ballscrews from two regular ballscrews instead of purchasing an already preloaded ballscrew. I'd like to keep it in English measurements if possible, but am open to metric if it really is a big price difference. So far I've been looking at 5/8 ballscrew with 5 revolutions per inch (.2 lead)

I'd like to keep the accuracy on the ballscrew to around .001 per foot, but if it is slightly more than that, it could be ok. So far the best deal I have found is for Nook XPR ballscrews at about 21 dollars per foot.

The real problem then lies with the ballnuts. At nook, the regular ballscrews are about 41 dollars EACH, and I need 2 per axis to preload them. Thats 246 dollars just in ballnuts.

Does anybody know of a decent company making ballscrews, with good accuracy and cheaper ballnuts?

Does anyone know if you can mix ballscrews of one company with ballnuts of another?

[HimyKabibble]

Originally Posted by firestorm1284 Hey everyone, still looking for more ballscrew info. I'm doing an X3 conversion and am definitely going to be using ballscrews. I want to keep the accuracy as high as reasonably possible. So far it seems that will involve a rolled ballscrew instead of a ground one for cost reasons, and also fabricating preloaded ballscrews from two regular ballscrews instead of purchasing an already preloaded ballscrew. I'd like to keep it in English measurements if possible, but am open to metric if it really is a big price difference. So far I've been looking at 5/8 ballscrew with 5 revolutions per inch (.2 lead) I'd like to keep the accuracy on the ballscrew to around .001 per foot, but if it is slightly more than that, it could be ok. So far the best deal I have found is for Nook XPR ballscrews at about 21 dollars per foot. The real problem then lies with the ballnuts. At nook, the regular ballscrews are about 41 dollars EACH, and I need 2 per axis to preload them. Thats 246 dollars just in ballnuts. Does anybody know of a decent company making ballscrews, with good accuracy and cheaper ballnuts? Does anyone know if you can mix ballscrews of one company with ballnuts of another?

The Nook XPR screws are the only rolled screws I'm aware of rated at 0.001"/foot. Most others are 0.003-0.004"/foot, some even more. Where did you get your pricing? That's close to what I paid directly from Nook for 1" XPR screws and nuts barely over a year ago.

You really shouldn't mix and match across manufacturers, as there's no way of knowing if the raceway profile of one is the same as another, and if it isn't you're very likely to see reduced performance and/or lifetime.

Regards,
Ray L.

[firestorm1284]

The quote was from Nook, they must have raised their prices! By the way I read your posts with caleb105 and saw your setup and how you preloaded your nuts. It looks like a sweet setup! I'd like to do something similar if I can do it without breaking the bank.

Well, I guess your answer begs the question.. is .003-.004"/foot acceptable if you preload your ballnuts and don't have any backlash in the rest of the system?

My understanding is that the ballnuts from nook have a backlash of .01, but you can increase ball size to get it down to about .003 at the cost of increased wear on the screw. Is .003 an acceptable amount of backlash? How much faster will my ballscrews wear with the larger balls?

I guess what I'm driving at is: if I don't want to spend 374 dollars for the nook screw and ballnuts, is there an OK option that will give me decent accuracy at a decent savings?

As an example, I see Roton has ballnuts at about 23 dollars each, and ballscrews at 10 dollars per foot, which would then run me only 198 dollars for the whole setup. But I contacted Roton and their ballscrews only are accurate to .009-.012. That is unacceptable, but is there some other company that may have better accuracy, but still with a substantial savings.

[HimyKabibble]

Originally Posted by firestorm1284 The quote was from Nook, they must have raised their prices! By the way I read your posts with caleb105 and saw your setup and how you preloaded your nuts. It looks like a sweet setup! I'd like to do something similar if I can do it without breaking the bank. Well, I guess your answer begs the question.. is .003-.004"/foot acceptable if you preload your ballnuts and don't have any backlash in the rest of the system? My understanding is that the ballnuts from nook have a backlash of .01, but you can increase ball size to get it down to about .003 at the cost of increased wear on the screw. Is .003 an acceptable amount of backlash? How much faster will my ballscrews wear with the larger balls? I guess what I'm driving at is: if I don't want to spend 374 dollars for the nook screw and ballnuts, is there an OK option that will give me decent accuracy at a decent savings? As an example, I see Roton has ballnuts at about 23 dollars each, and ballscrews at 10 dollars per foot, which would then run me only 198 dollars for the whole setup. But I contacted Roton and their ballscrews only are accurate to .009-.012. That is unacceptable, but is there some other company that may have better accuracy, but still with a substantial savings.

Personally, I think most people worry WAY too much about the "paper" accuracy of their machines. Getting a machine that is truly accurate to 0.001" takes a LOT of work, and a LOT of money. You'll have to stiffen up all kinds of things, replace ALL the bearings, and all kinds of things. For 99.99% of what you'll do, you simply don't NEED that accuracy. For the remaining 0.01%, you can get there by sneaking up on the final dimension.

As for how much backlash is too much - That depends on your expectations. Even 0.001" of backlash *will* leave visible artifacts on your work. But, so will flex in the machine, flex in the tool, etc., etc.. If you're expecting this thing to give you the same result as a $50K VMC, you need to reset your expectations to something more reasonable. Your parts *will* require some cleanup after coming off the machine.

Re: the cost of the Nook screws/nuts: You're talking about a difference of what - maybe $100-200 in the total cost of your conversion to use the Nook parts. You'll have WELL over $2K in the machine by the time you're done. Does it make sense to compromise the quality in such an important area just to save such a small amount of money?

Regards,
Ray L.

[firestorm1284]

Thanks for the info Ray, that's about what I expected the reply would be. The problem is that it seems to be a recurring theme here: JUST a few hundred dollars more for this and JUST a few hundred more for that. It's adding up!

Well I guess I'll go with the Nook screws and nuts! Thanks for the input!

[escott76]

Originally Posted by firestorm1284 The quote was from Nook, they must have raised their prices! By the way I read your posts with caleb105 and saw your setup and how you preloaded your nuts. It looks like a sweet setup! I'd like to do something similar if I can do it without breaking the bank. Well, I guess your answer begs the question.. is .003-.004"/foot acceptable if you preload your ballnuts and don't have any backlash in the rest of the system? My understanding is that the ballnuts from nook have a backlash of .01, but you can increase ball size to get it down to about .003 at the cost of increased wear on the screw. Is .003 an acceptable amount of backlash? How much faster will my ballscrews wear with the larger balls? I guess what I'm driving at is: if I don't want to spend 374 dollars for the nook screw and ballnuts, is there an OK option that will give me decent accuracy at a decent savings? As an example, I see Roton has ballnuts at about 23 dollars each, and ballscrews at 10 dollars per foot, which would then run me only 198 dollars for the whole setup. But I contacted Roton and their ballscrews only are accurate to .009-.012. That is unacceptable, but is there some other company that may have better accuracy, but still with a substantial savings.

I used Thompson screws and nuts from McMaster. $27 each for the nuts, and ~$10 a foot screws. After all was said and done with the washers and buying a tap to make the ball nut holders I figure I was at about $100 per axis.
The screws I used had a max lead error in the .004" a foot range. With pre-loaded nuts and that small an error, If I can't design parts to deal with that tollerance I'm in trouble. It's also a repeatable error, so it's even less of a worry.
As was said earlier, it does seem that people make a huge deal out of trying to eliminate everything in the screw and nut. I did best effort across all sources I could come up with, but I accept that there is only so good I'm reasonably going to get. Don't forget how the screw is held in what bearings will make a difference in backlash as well.

[firestorm1284]

Thanks for the great info again! Escott, What website did you actually BUY your screws and nuts from? I tried searching for Thompson Saginaw and I can't find their website. That sounds like a reasonable accuracy for the price..

[escott76]

Originally Posted by firestorm1284 Thanks for the great info again! Escott, What website did you actually BUY your screws and nuts from? I tried searching for Thompson Saginaw and I can't find their website. That sounds like a reasonable accuracy for the price..

McMaster.com part numbers 5966K26 for the screw and 5966K16 for the nuts.